image classification
The task of assigning labels to images based on their content. In AI, this is typically achieved using convolutional neural networks (CNNs) that learn to identify and categorize objects or features present in the images.
- A Stable Whitening Optimizer for Efficient Neural Network Training
- AdaMSS: Adaptive Multi-Subspace Approach for Parameter-Efficient Fine-Tuning
- Alias-Free ViT: Fractional Shift Invariance via Linear Attention
- Bipolar Self-attention for Spiking Transformers
- CAT: Content-Adaptive Image Tokenization
- Certifying Deep Network Risks and Individual Predictions with PAC-Bayes Loss via Localized Priors
- Class conditional conformal prediction for multiple inputs by p-value aggregation
- Conformal Inference under High-Dimensional Covariate Shifts via Likelihood-Ratio Regularization
- Correlated Low-Rank Adaptation for ConvNets
- DAMamba: Vision State Space Model with Dynamic Adaptive Scan
- DualOptim: Enhancing Efficacy and Stability in Machine Unlearning with Dual Optimizers
- FairDD: Fair Dataset Distillation
- GSPN-2: Efficient Parallel Sequence Modeling
- Gatekeeper: Improving Model Cascades Through Confidence Tuning
- Generalized Gradient Norm Clipping & Non-Euclidean $(L_0,L_1)$-Smoothness
- Generalized Gradient Norm Clipping & Non-Euclidean $(L_0,L_1)$-Smoothness
- Gompertz Linear Units: Leveraging Asymmetry for Enhanced Learning Dynamics
- H-SPLID: HSIC-based Saliency Preserving Latent Information Decomposition
- Head Pursuit: Probing Attention Specialization in Multimodal Transformers
- KOALA++: Efficient Kalman-Based Optimization with Gradient-Covariance Products
- Knowledge Distillation Detection for Open-weights Models
- MaxSup: Overcoming Representation Collapse in Label Smoothing
- MaxSup: Overcoming Representation Collapse in Label Smoothing
- Mind the Gap: Removing the Discretization Gap in Differentiable Logic Gate Networks
- MobileODE: An Extra Lightweight Network
- Modelling the control of offline processing with reinforcement learning
- Neural Tangent Knowledge Distillation for Optical Convolutional Networks
- Optimal Control for Transformer Architectures: Enhancing Generalization, Robustness and Efficiency
- PROFIT: A Specialized Optimizer for Deep Fine Tuning
- Parameter Efficient Fine-tuning via Explained Variance Adaptation
- QuadEnhancer: Leveraging Quadratic Transformations to Enhance Deep Neural Networks
- REP: Resource-Efficient Prompting for Rehearsal-Free Continual Learning
- SDPGO: Efficient Self-Distillation Training Meets Proximal Gradient Optimization
- SketchMind: A Multi-Agent Cognitive Framework for Assessing Student-Drawn Scientific Sketches
- Spectral Graph Neural Networks are Incomplete on Graphs with a Simple Spectrum
- To Think or Not To Think: A Study of Thinking in Rule-Based Visual Reinforcement Fine-Tuning
- Toward Relative Positional Encoding in Spiking Transformers
- Weak-to-Strong Generalization under Distribution Shifts
- When majority rules, minority loses: bias amplification of gradient descent